← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
1k
tokens
903.7
PP tok/s
88.7
TG tok/s
1133
TTFT (ms)
20.1
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 903.7 | 88.7 | 20.1 GB | current |
| 4k | 915.9 | 83.0 | 20.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 88.7 | 1.00× |
| 2× | 133.3 | 1.50× |
| 4× | 166.8 | 1.88× |
| 8× | 201.1 | 2.27× |